[llvm] [AMDGPU] Automatic conversion from wave32 to wave64 (PR #137376)

[via llvm-commits]

================
@@ -0,0 +1,121 @@
+; RUN: opt -S -mtriple=amdgcn -mcpu=gfx1100 -passes=si-convert-wave-size < %s | FileCheck %s
----------------
alex-t wrote:

There are both kind of tests: test_barrier and test_non_convergent are positive but the rest are negative. They check that we're not converting kernels that do not meet the requirements listed above. Is this what you meant?

https://github.com/llvm/llvm-project/pull/137376